5.0 The Small Finds

Introduction | Overview of the assemblage | The distribution of the finds | The military equipment and its implications | Personal ornaments | Military equipment | Writing equipment | Household | Fasteners and fittings | Recreation | Craft and industry | Structural finds | Miscellaneous | Coins | Report on flint | Geological report on the querns | Slag | Geological report on the hones | Ceramic building material | The structural stone

by H.E.M. Cool, with contributions by C. Barclay and G. Woolrich

5.1 Introduction

This section deals with all of the small finds other than the flint. It includes all those items found in contexts that can be stratigraphically assigned to the Roman period, together with those artefacts interpreted as Roman on typological grounds that were found in non-Roman contexts. In total there are 343 items that fall into this category. The stratified pieces are summarised in Table 15 by material and period group. In this table the material from the evaluation trenches (WRB91) has been divided into two. Those items from late dumps that appear to be the equivalent of context group 1.12 in the main excavation have been included in the context group 1.12 total. The rest of the WRB91 items from Roman contexts are given in the bottom row of the table without further sub-division. As will be seen from Table 15, the assemblage is dominated by iron nails and other iron items and there was no preservation of organic finds other than skeletal material, though organic material is preserved as replacement products in iron corrosion in two cases (see nos.37 and 72).

Period Non-ferrous Iron Nail Glass object Glass vessel Bone, antler, shell Stone Pottery Total
0 1 - 1 - 1 --- 3
1.1- 1 1 ----- 2
2.1----- 1 -- 1
3.1- 1 6 ----- 7
3.2- 1 3 - 3 --- 7
4.1 1 3 7 - 3 --- 14
4.3 1 3 2 --- 1 - 7
4.4- 1 2 ----- 3
4.5- 9 38 1 1 --- 49
4.6- 1 3 --- 2 - 6
4.7 1 - 5 - 1 1 -- 8
4.8 1 - 4 ----- 5
5.1 2 8 44 1 3 3 -- 61
5.2-- 1 - 1 --- 2
6.11 28 79 -8 1 4 4 125
Total 8 56 196 2 21 6 7 4 300
WRB 91- 4 11 - 3 1--19
Table 15: Distribution of small finds according to period

The section is structured as follows. First of all there is an overview of the assemblage according to functional category and date. This contains links to the detailed typological section where the principal items are catalogued, illustrated and identified. The electronic publication of this report has enabled the presentation of material that would normally be assigned to the archive. The archive pages present, in the main, material that is not suitable for illustration such as the bulk of the nails, the unidentified fragments of metalwork, the body fragments of glass vessels etc. The entries in the typological section provide links to these pages where appropriate. Following the overview, the distribution of the finds across the site is considered. Finally the implications of the military metalwork are discussed.

5.2 Overview of the assemblage

One aspect of the assemblage consists of personal items such as pieces of jewelry (nos.1-8), hobnails from shoes (no.9), military equipment (nos.10-11) and writing equipment (no.12). There are also household items such as glass vessels (nos.13-23), an uncommon iron vessel (no. 24) and possible iron fittings from household utensils (nos.25-6). Knives and handles (nos.27-9) and quernstones (nos.30-31) also fall into the general household category. The commonest items in the fasteners and fittings category are the keys (nos.32-4) but there is also a barb-spring padlock bolt (no.35) and a short length of chain (no.36).

Recreation may be indicated by a group of re-used pottery roundels (nos.38-44). The evidence for craft and industry is relatively slight but there is evidence for the activities of a blacksmith (nos.45-6), a carpenter (no.47) and someone working antler (nos.48-9). The hones (nos.50-2) might either be associated with craft or general household purposes. Other than the single fragment of window glass (no.53), all the structural finds are made of iron and include possible wall hooks or pivots (nos.54-5), double-spiked loops and staples (nos.56-8), joiner's dogs (nos.59-60), nails (nos.61-2) and a holdfast (no.63). All-purpose items such as rings and fragmentary items whose origins cannot be identified are also present (nos.64-77). In addition to these finds there is also one associated with a burial (a casket or box no.37).

Much of this material is not chronologically sensitive and cannot be more closely dated than to say it is Roman. As will be appreciated, much of the assemblage is made up of functional items such as hobnails, tools and structural finds and these do not change with the whims of fashion. The most chronologically sensitive items are the personal ornaments, the military equipment and the vessel glass. The only undoubted item of 1st century date is a Hod Hill brooch (no.1) though the possible iron penannular brooch (no.3) and the glass bangle (no.6) could be contemporary and in the case of the glass bangle is most likely to be of Trajanic date or earlier. The vessel glass can be divided into three broad dating groups. The jugs (nos.16-18) belonging to the later 1st to mid 2nd century, the bottles of the later 1st to early to mid 3rd century (nos.19-23) and the drinking vessels the later 2nd to mid to late 3rd century (nos.13-15). With the exception of a single blue/green cylindrical bottle fragment (no.22) which is of later 1st or early 2nd century date, and a hemispherical cup (no.16) which may be as late as the later 3rd century, all of the vessel glass could be dated to the mid Antonine to Severan period. It is to this period that the military equipment (nos.10-11) also belongs. Of the independently dated material the hemispherical cup from a modern context (no.16) is the latest item in the assemblage. There is no material that must be of 4th century date.

5.3 The distribution of the finds

As can be seen from Table 15 over half the stratified finds were associated with the late 3rd/early 4th century dumping of context groups 1.11 and 1.12. The dumped material included a wide range of finds including a hair pin (no.4), at least one shoe sole (no.9), both items of military equipment (nos.10-11), a stylus (no.12), vessel glass (nos.20-1, 23), an iron bowl (no.24), a possible suspension fitting (no.26), knife blades (nos.27-8), quernstones (no.31), a key (no.33), a chain fragment (no.36), counters (nos.38, 40-1, 43-4), tools (nos.45-7), antler waste (nos.48-9), a hone (no.51) and structural fittings (nos.54, 56, 59, 61-3). This range of finds suggests the dumps were incorporating material from general rubbish disposal and possibly the clearance of structures. The more closely datable items, such as the military equipment and the vessel glass, suggest the rubbish may have been accumulating as early as the late 2nd century. As with any deliberately dumped material, the relationship of the material in the dump to its immediate surroundings is open to question. Though it may have been derived from activities in the immediate surroundings it could as easily have been brought in from elsewhere.

The other main group of finds in Trench 1 came from the pits and post-holes of Building D (Period 4.5; group 1.31). The finds recovered tend to be less diagnostic than those from the dump. However, they include a fragment of colourless vessel glass, most likely of 2nd or 3rd century date; a group of hobnails (nos.740, 742, 744, 745, 747), nails and other fragments of iron; and a glass bead (no.7), though the last-mentioned may be a modern intrusion.

In Trench 2 the biggest concentration of finds came not from dumps but from contexts associated with the construction and use of Building A (Period 5.1; groups 2.8-2.11). These include a hairpin or needle fragment (no.5), a glass bead (no.8), hobnails, vessel glass (no.14), a barb spring padlock bolt (no.35) and of course the ubiquitous iron nails. This range suggests general domestic activity and does not hint at any specialised use. The only closely datable item is the fragment of vessel glass, which comes from the type of cup common in the late 2nd to early 3rd century.

5.4 The military equipment and its implications

Perhaps the most interesting aspect of the finds assemblage from Welton Road is the military metalwork (nos.10-11), found in the late 3rd/early 4th century dump. Both items may be dated to the later 2nd or 3rd century and clearly cannot be associated with the Flavian to early Hadrianic military occupation. These are not the only military items of the late 2nd to 3rd century date at Brough-on-Humber as a hexagonal mount with an oval boss was found during the Wacher excavations (Wacher 1969, 89 no.19, fig.38). Such mounts came into use during the late 2nd century and are commoner in the 3rd century (Oldenstein 1977, 139, 248 nos.267-72, Taf. 34). A rosette mount with rectangular loop at the back might also fall into this category (Corder 1935, 32 no.8, fig.8). It is not a well-established military type, but such a rosette head occurs on a variety of military fittings of this date (see for example Oldenstein 1977, Taf. 46 nos.483-4, 57 nos.704-8, 62 nos.797-8).

Although the military origins of Brough-on-Humber have never been disputed, its status following the evacuation of the fort early in the Hadrianic period has been much debated (see Section 1.2). These finds strongly suggest a military presence at some time during the later 2nd or 3rd century. They cannot be used alone to assert that Brough was primarily a military establishment as similar equipment has been found on apparently civilian sites such as the civitas capital at Aldborough (Bishop 1996, 67-73 nos.422-48, figs.37-9) and the villa at Dalton Parlours, West Yorkshire (Cool 1990, 81). However, this equipment is useful in providing definite evidence of the presence of soldiers at Brough at the time when the defences were being built in the early 3rd century. It has been pointed out that the town defences appear to follow a military model (Wacher 1969, 32), and these items of military equipment suggest they could have been built by military personnel. This must strengthen Wacher's contention that Brough stands out as being completely different from the other provincial civitas capitals (Wacher 1995, 398).


© Internet Archaeology URL:
Last updated: Tue Nov 28 2000